Ontologies & Its Applications Ontologies & Its Applications - PowerPoint PPT Presentation

Ontologies & Its Applications Ontologies & Its Applications San Su Lee, Jong Lim, Rami Al-Ghanmi San Su Lee, Jong Lim, Rami Al-Ghanmi

Outline Outline � Introduction to Ontologies � Introduction to Ontologies � Definition � Definition � Web Ontology Language (OWL) � Web Ontology Language (OWL) � Ontology Generation from Tag Spaces � Ontology Generation from Tag Spaces � The Problem � The Problem � Tag Meta-Data � Tag Meta-Data � Ontology Clusters � Ontology Clusters � Ontology-Based Federation of Data � Ontology-Based Federation of Data � Data Types � Data Types � System Architecture � System Architecture

Introduction to Ontologies Introduction to Ontologies

Ontology Ontology � Definition � Definition � An ontology is a specification of conceptualization � An ontology is a specification of conceptualization � Practical Used as a form of Knowledge Representation � Practical Used as a form of Knowledge Representation � Semantic Web, Software Engineering, Artificial Intelligence, � Semantic Web, Software Engineering, Artificial Intelligence, Information Architecture Information Architecture � Taxonomy � Taxonomy � A Simple Ontology � A Simple Ontology � Differences between a classification and an ontology � Differences between a classification and an ontology � The richness of information available � The richness of information available

Elements of an Ontology 1 Elements of an Ontology 1 � Individuals � Individuals � Instances � Instances � Ground level component of an ontology � Ground level component of an ontology � Concrete Object � Concrete Object � People, Animal, Planets � People, Animal, Planets � Abstract Object � Abstract Object � Numbers, Words � Numbers, Words

Elements of an Ontology 2 Elements of an Ontology 2 � Classes (Concepts) � Classes (Concepts) � Abstract groups, sets, or collection of objects � Abstract groups, sets, or collection of objects � Contain individuals, other classes or combination of both � Contain individuals, other classes or combination of both � Person: the class of all People � Person: the class of all People General Class Vehicle Car Truck 2-Wheel 4-Wheel Specific Class Drive Drive

Elements of an Ontology 3 Elements of an Ontology 3 � Attributes � Attributes � At least a Name and a Value � At least a Name and a Value � A complex data type � A complex data type � Example � Example � Object: a Ford Explorer � Object: a Ford Explorer � Attributes: � Attributes: � Name: Ford Explorer � Name: Ford Explorer � Number of Doors: 4 � Number of Doors: 4 4-Wheel � Engine: { 4.0L, 4.6L} � Engine: { 4.0L, 4.6L} Drive � Transmission: 6 speed � Transmission: 6 speed Ford Explorer

Elements of an Ontology 4 Elements of an Ontology 4 � Relationships � Relationships � Important type of relation is the subsumption: � Important type of relation is the subsumption: � is-subclass-of, the converse of is-a, is-type-of, is-subclass-of � is-subclass-of, the converse of is-a, is-type-of, is-subclass-of � Example � Example � Object: Ford Branco � Object: Ford Branco � Attribute: Successor: Ford Explorer � Attribute: Successor: Ford Explorer Car 2-Wheel 4-Wheel Drive Drive Ford Ford Bronco Explorer

Why Ontologies? Why Ontologies? � Sharing common understanding of the structure of � Sharing common understanding of the structure of information among people or software agents information among people or software agents � Reusing of domain knowledge � Reusing of domain knowledge � Making domain assumptions explicit � Making domain assumptions explicit � Separating domain knowledge from the operational � Separating domain knowledge from the operational knowledge knowledge � Analyzing domain knowledge � Analyzing domain knowledge

Ontology Languages Ontology Languages � Web Ontology Language (OWL) � Web Ontology Language (OWL) � A formal language used to encode the ontology � A formal language used to encode the ontology � To process the content of information instead of � To process the content of information instead of presenting information presenting information � Supported by � Supported by � XML: provides a surface syntax for structured documents (no � XML: provides a surface syntax for structured documents (no semantic constraints) semantic constraints) � XML Schema: Restricting the structure of XML document � XML Schema: Restricting the structure of XML document � RDF: A data model for objects and relations between them � RDF: A data model for objects and relations between them � RDFS: A vocabulary for describing properties and classes of RDF � RDFS: A vocabulary for describing properties and classes of RDF resources resources

OWL Sublanguages OWL Sublanguages � OWL Lite � OWL Lite � Support a classification hierarchy and simple constraints � Support a classification hierarchy and simple constraints � A quick migration path for thesauri and other taxonomies � A quick migration path for thesauri and other taxonomies � Lower formal complexity � Lower formal complexity � OWL DL � OWL DL � Support the maximum expressiveness while retaining � Support the maximum expressiveness while retaining computational completeness and decidability computational completeness and decidability � Including all OWL language constructs, Using only under � Including all OWL language constructs, Using only under certain restrictions certain restrictions � OWL Full � OWL Full � Maximum expressiveness and the syntactic freedom of � Maximum expressiveness and the syntactic freedom of RDF with no computational guarentees RDF with no computational guarentees

Published Ontologies Published Ontologies � Dublin Core � Dublin Core � A simple ontology for documents and publishing � A simple ontology for documents and publishing � WordNet � WordNet � Lexical reference system � Lexical reference system � Gene � Gene � Ontology for genomics � Ontology for genomics � SBO � SBO � Systems Biology Ontology for computational model in biology � Systems Biology Ontology for computational model in biology � LinkBase � LinkBase � A formal representation of the biomedical domain, founded on BFO � A formal representation of the biomedical domain, founded on BFO (Basic Formal Ontology) (Basic Formal Ontology) � FOAF Friend-of-a-Friend � FOAF Friend-of-a-Friend

Ontology Generation from Tag Spaces Ontology Generation from Tag Spaces

Tags Tags � A relevant keyword or term associated with or � A relevant keyword or term associated with or assigned to a piece of information assigned to a piece of information � Describes the item and enabling keyword-based � Describes the item and enabling keyword-based classification of information it is applied to classification of information it is applied to � is usually chosen informally and personally by the � is usually chosen informally and personally by the author/creator or the consumer of the item author/creator or the consumer of the item

Problem: searching the TagSpace Problem: searching the TagSpace How would How would You tag this? You search For it? Tags: Ikura, Uni, Ebi, Sushi, Nigiri, Japanese food, lunch in Tokyo, Ezobafun-uni, Kitamurashiuni, Murasakiuni, Akazaebi, Tenagaebi, etc.

Problem: exploring the TagSpace Problem: exploring the TagSpace Not usable !

What is Missing ..? What is Missing ..? � “Tag Relations improve searchability and exploration.” � “Tag Relations improve searchability and exploration.” � Similar Tags � Similar Tags � Spelling and morphology � Spelling and morphology � macos< -> mac_os< -> mac os; tagging < -> tags < -> tagged, � macos< -> mac_os< -> mac os; tagging < -> tags < -> tagged, � Synonyms: � Synonyms: � macos < -> tiger; films < -> movies; new york < -> nyc; � macos < -> tiger; films < -> movies; new york < -> nyc; � Related: � Related: � cooking < -> recipes, software development < -> programming, � cooking < -> recipes, software development < -> programming, � Tag Groups or Subtags � Tag Groups or Subtags � Location -> san francisco, london, new york, etc. � Location -> san francisco, london, new york, etc. � Food -> sushi, sashimi, pizza, etc. � Food -> sushi, sashimi, pizza, etc. � Programming -> html, java, css, etc. � Programming -> html, java, css, etc.

1. Get Meta Data 1. Get Meta Data

2. Build Tag Relation Graphs 2. Build Tag Relation Graphs

Compute Similarity Compute Similarity

Ontology Cluster Ontology Cluster

Ontology-Based Federation of Data Ontology-Based Federation of Data

The Problem The Problem � Representation of Geo-scientific Data � Representation of Geo-scientific Data � Different Data Sources � Different Data Sources � Different Data Representations � Different Data Representations � Different Data Types � Different Data Types � Facilitate the use of this data � Facilitate the use of this data